Abstract: Normally data cube is very large and relations among cells are very complicated. So semantic data cube is difficult to realize. Based on quotient cube, Semantic Data Cube(SDC) structure is put forward in this paper. Each cell in lattice expressed by its upper bound and low bound is also preserved. SDC depicts the lattice of cells concisely and stores data cube compactly and can keep all the semantic relations. The operators of drill-down and roll-up between cells can be done in SDC. Applying semantics to answer query and maintain incrementally, the cost of queries and updating can be reduced greatly. Experimental results show that SDC is effective.

Key words: data warehouse, data cube, semantics, incremental maintenance

摘要： 通常数据立方体体积较大，语义关系复杂，完整的语义立方体很难实现。基于商立方体，该文提出了语义数据立方体结构(SDC)，将单元格中的单元以其上界替代，并保存下界，简化了单元格的表示，保持单元格的全部语义，并可以实现单元的上卷和下钻操作。把语义关系应用到数据立方体的查询、增量更新中，使查询响应时间及更新代价大大降低。实验结果表明，SDC是有效的。

关键词: 数据仓库, 数据立方体, 语义, 增量维护

CLC Number: 

• TP391